| Abstract | This paper uses the province level panel data from China to investigate the scale of corruption in China provinces during 2001 to 2011. A multiple indicators and multiple causes (MIMIC) model based on the latent variable structural theory have been applied. The model includes government size, and strength of anti-corruption, citizen education level, fiscal decentralization, public procurement and public officials’ relative wage as courses of corruption, and takes the GDP growth rate, income inequality, public investment as indicators of corruption. The coefficients for the government size is found to be positive and significant. The coefficients for the strength of anti-corruption, citizen education level, fiscal decentralization, public procurement and public officials’ relative wage are found to be negative and significant. Furthermore, the coefficients for corruption with income inequality and public investment are positive and significant, while the coefficients for corruption with economic growth are negative and significant, which means with the rise of corruption scale, leading to a much bigger investment and income inequality while inhibiting economic growth rate. The empirical results show that, the national average corruption index during 2001-2011 is gradually downward trend. Besides, the east, middle and west regions of the corruption index is also gradually downward; the east region of corruption is the lowest, which is lower than the national average level of corruption as well as the middle region, while the west region of the corruption index is the highest, much higher than the national average of corruption index. |